Click for next page ( 27


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 26
Characterizing Exposure ... Final Report PROJECT 2: COVARIATES, CONFOUNDERS, AND CONSISTENCY: CHARACTERIZING THE VIETNAM VETERAN FOR EPIDEMIOLOGIC STUDIES The Columbia University researchers' proposal indicated that the primary aim of Project 2 was to identify and evaluate key variables and other aspects of military experience in Vietnam, such as herbicide handling and exposure to combat stress, that need to be taken into account in future epidemiologic studies of Vietnam veterans. Such knowledge will be essential in developing the conceptual framework and design of future studies. A second, equally important goal was to examine and evaluate the quality of methods of military-unit identification for individual veterans, because it is through the military unit that linkage with the herbicide-exposure database is effected. One aspect of reaching this goal was to assess the reliability of veterans' recall and of various approaches to obtaining military-unit data on people who may be enrolled in epidemiologic studies. A third goal was to address issues of generalizability by carrying out surveys in different populations of veterans with a focus on assessment of exposure to herbicides, military combat, and physical and mental-health outcomes that are likely to be relevant to future Vietnam-veteran studies. To accomplish those objectives' . . the researchers designed and conducted three substudies: A follow-up study of members of the American Legion who were originally surveyed in 1984-1985 and who made up the study population in previously published epidemiologic work (Stellman SD et al., 1988a, 198Sb; Stellman IM et al., 1988; Snow et al., 1988), A survey of veterans in minority groups (blacks and Latinos) that may have been underrepresented in previous studies, 26

OCR for page 26
Characterizing Exposure ... Final Report A survey of a population of women both veterans and civilians who were stationed in Vietnam during the conflict and on whom few data are available. Follow-Up of the American Legion Survey Enduring Effects of Vietnam Service The researchers recovered the original data files and identified the earlier cohort of 12,748 men who were randomly sampled from the American Legion membership for a 1984 survey (the Wave 1 survey). The American Legion membership department was able to trace about 60% of them from current membership roles. An additional 12% were then located by attending Legion state conventions, networking with attendees, and circulating lists of"missing" Legionnaires to individual legion posts (organizational units). Some remaining veterans were found through on- line searches and use of a commercial credit-rating service that maintains an extensive name and address database. Of the 72% found, a follow-up determined that some were deceased and a small number were not eligible for participation because their service did not meet the study requirements. In total, 63% of the original cohort was sent questionnaires. A survey nearly identical with that used in 1984 was reviewed and slightly amended after discussions with VA and American Legion officials, subjected to institutional review board approval at Columbia University, and mailed to nearly 9,000 Legionnaires in 1998 (the Wave 2 survey). Investigation of the data on respondent characteristics and the response rate yielded some interesting results that are potentially important for future epidemiologic studies. Those with the highest posttraumatic stress disorder (PTSD) scores and those in the poorest health were much more likely to be lost to followup. It has been found that PTSD is highly correlated with combat and combat with Agent Orange exposure scores, so special care must be taken in the design of study populations and the execution of epidemiologic studies to ensure that one does not miss precisely the groups one is most interested in studying. Papers addressing the survey methods and results in greater detail are listed in Appendix A. 27

OCR for page 26
Characterizing Exposure ... Final Report Assessing the Reliability of Veterans' Recall Among the items assessed in the survey discussed above were the test-retest reliability and internal consistency of responses. Both demographic and health-outcome variables were evaluated. Earlier work by the researchers had found remarkable consistency in reports of combat experience in this cohort (Stellman SD et al., 1988a). In the present work, they found that respondents were largely consistent in self-reports of physical and mental health between the two waves of the survey. That suggests that surveys are a valid means of gathering data from this population. Outreach to Black and Latin o Veterans Although it represents a wide range of educational and income levels, the original American Legion cohort is predominantly white, so there were concerns over the generalizability of results. Furthermore, the National Vietnam Veterans Readjustment Study (NVVRS) suggested that black and Latino veterans are particularly at risk for negative outcomes related to service in Southeast Asia (Kulka et al., 1990; Stellman SD et al, 1988a). Given the documented association between combat and Agent Orange exposure, that suggests that veterans from those groups might have been at higher risk for exposure to herbicides than white veterans (Buliman et al., 1990; Breslin et al., 19883. The investigators consulted with the American Legion leadership and identified several posts in Ohio and Maryland that participated in the original study and had large numbers of black members. In addition, they identified posts in New Mexico and Texas with large numbers of Hispanic members and enlisted the cooperation of the American Legion in Puerto Rico. Questions regarding perceptions of discrimination in the military were adapted from those included in the NVVRS and were added to the survey. The wording of the questions was changed so that respondents of any race or ethnic group could answer them in as much as none of the posts in the outreach was segregated by race or ethnicity and white veterans would thus be expected among the respondents. The outreach sample comprised 1,230 respondents from all the 28

OCR for page 26
Characterizing Exposure ... Final Report regions listed above. Approximately 26% of the cohort iclentified themselves as black; about 52% as Hispanic (white). The results of the effort are describer! in the contractor's final report (Steliman and StelIman, 2003~. The Columbia University researchers have preparer! papers intended for publication in peer- reviewed journals regarding both the enduring effect of combat exposure with respect to PTSD ant! the more general health and well-being effects of combat exposure among American Legion members. They assessed the validity of veteran recall in this longitudinal cohort and have found a high level of agreement between questionnaires responses from 1984 and 1998. The work is described in their final report (Stellman and Stellman, 2003~. Study of Women Veterans In response to the committee's request that the researchers study the generalizability of their survey and conclusions, the researchers also constructed a study of women veterans. Women shared some stressors with men but were also exposed to unique stressors related either to their occupational assignments or to their sex, and these stressors could have had a lasting effect on their physical and mental health. The researchers were provided with an unexpected opportunity by the Vietnam Women's Memorial Foundation (VWMF)~7 to use its membership database to catty out an extensive study of nearly 2,500 military and civilian women who served in Vietnam. To obtain wartime location data on the respondents, a comprehensive list was developed containing military units in which large numbers of women were known to have served; a large proportion of women veterans served as nurses or in other medical roles' so many of the locations were hospitals and other medical facilities. The researchers also used focus groups to ensure that they covered subjects of relevance and concern to women veterans, and they included questions that were relevant to the experiences of nonmilitary women who served with the Red Cross and other nongovernment |7 VWMF (previously called the Vietnam Women's Memorial Project) is a nonprofit organization that, among other missions, is seeking to identify the military and civilian women who served during the Vietnam War, to educate the public about their role, and to facilitate research on the physiologic, psychological, and sociologic issues correlated to their service. 29

OCR for page 26
Characterizing Exposure ... Final Report organizations. As a result, a number of changes were made in the American Legion questionnaire to accommodate the experiences of women in Vietnam; for example, the researchers expanded coverage of issues related to sexual harassment. In brief, the researchers found that previously developed measures of stressors (Wolfe et al., 1993) were reliable for women who served in Vietnam and that future epidemiologic studies of women veterans need to take such measures into account when assessing health risks related to their service in Vietnam. The results are discussed in the contractor's final report (StelIman and StelIman, 2003) and will be addressed in future papers to be submitted to peer-reviewed journals. Continuing Work In addition to the papers cited above, the researchers indicate that they have several manuscripts regarding topics related to covariates, confounders and consistency in preparation or in early stages of submission. They will elaborate on information contained in the final report and disseminate it in the peer-reviewed literature. Accomplishments Design and implementation of three studies examining self-reports of military service, physical and mental health, and other potential confounders and covariates of methodologic interest (for example, combat experience): a followup survey of a large cohort of Vietnam veterans who are members of the American Legion, a new survey of black and Latino Legionnaires who are veterans of Vietnam, and a new study of women veterans and civilians who were stationed in Vietnam during the conflict. . An empirical evaluation of the validity of veterans' recall. 30